Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
Int. J. Electrochem. Sci., 15 (2020) 6109 – 6121, doi: 10.20964/2020.07.75
International Journal of
ELECTROCHEMICAL SCIENCE
www.electrochemsci.org
Electrochemical Evaluation of Titanium Production from
Porous Ti2O3 in LiCl-KCl-Li2O Eutectic Melt
Kun Zhao1,*, Feng Gao2
1 College of Material Science and Engineering, Yangtze Normal University, Chongqing 408100, China 2 School of Chemistry and Chemistry Engineering, Yangtze Normal University, Chongqing 408100,
China *E-mail: [email protected]
Received: 4 March 2020 / Accepted: 5 May 2020 / Published: 10 June 2020
This work presents an electrochemical co-reduction of porous Ti2O3 in LiCl-KCl-Li2O melt at 800 °C.
Cyclic voltammetry was applied to investigate the electrochemical behavior of Li+, Cl–, O2– and CO32–
by using glassy carbon and Pt working electrodes, respectively. The electrode reaction mechanisms and
particle evolution principles in molten salt were discussed. Transient electrochemical techniques show
that Ti2O3 is reduced to Ti metal by a two-step mechanism involving the exchanges of one and two
electrons. Continually-varied current electrolysis was conducted using a porous Ti2O3 electrode against
graphite and Pt anode in the molten LiCl-KCl-Li2O electrolyte medium at 800 °C respectively to gain
insight into the electro-reduction mechanism. The results showed the oxide component was electro-
deoxidized effectively and Ti metal was produced as the final product in the cathode. Our work suggests
that developing a low-cost inert anode material is essential for the production of titanium by electrolysis
in the future.
Keywords: Titanium; Continually-varied current electrolysis; Porous electrode; Ti2O3
1. INTRODUCTION
The use of the electron as a reductant in electro-deoxidation for the preparation of titanium is
known as a technique relatively simple, inexpensive, and eco-friendly [1, 2]. As the most typical
representative, the FFC Cambridge process based on the direct electrochemical reduction of TiO2 in
molten CaCl2 has been reported to be successful in small trials and considered to be more promising to
replace the current titanium production process——The Kroll process [3, 4]. The main mechanism of
the FFC process is the ionization of cathode TiO2 in molten chlorides under the driving of the potential
that is lower than the thermodynamic decomposition potential of the electrolyte but sufficient to
decompose the metal oxide to the metal. The ionized oxygen passes into molten salt and reacts with
Int. J. Electrochem. Sci., Vol. 15, 2020
6110
graphite anode to evolve CO or CO2. The electro-deoxidation of TiO2 pellets in molten CaCl2 was
studied by Schwandt et al [5]. The results showed that the electrochemical reduction of TiO2 in molten
CaCl2 was achieved by the formation and decomposition of Ca-Ti-O compounds and titanium sub-
oxides. The reduction pathway to the sequence of reactions follows as TiO2 → Ti4O7 + CaTiO3 → Ti3O5
+ CaTiO3 → Ti2O3 + CaTiO3 → CaTi2O4 + TiO → Ti2O → Ti. Ho-Sup Shin et al. investigated the
electrochemical reduction of TiO2 powder in molten LiCl [6, 7]. The reduction pathway is TiO2 →
LiTi2O4 → LiTiO2 → TiO → Ti2O → Ti.
Most of the previous works are devoted to the production of Ti used TiO2 [8–12]. However,
compared with TiO2, titanium sub-oxides possess a better electrical conductivity at a certain temperature
due to the unique structure. Among these compounds, Ti2O3 is the lowest oxide before forming Ti-O-C
via carbothermal reduction, which means it can be produced easily [13]. Besides, when Ti2O3 serves as
the cathode above 650 °C, the electrolytic efficiency can be improved significantly through accelerating
the electrochemical process [14]. In this paper, the possibility of electrolysis using a porous Ti2O3
electrode in LiCl-KCl-Li2O melt as a new process of Ti production was investigated. Compared with
electrolysis using TiO2, this method shortened and simplified the reduction pathway to producing Ti.
2. MATERIALS AND METHODS
2.1 Raw materials
Anhydrous LiCl, KCl, Li2O and Li2CO3 (99%, analytical grade) purchased from Sinopharm
Chemical Reagent Company Limited were used for electrochemical tests and electrolysis process in this
work. LiCl, KCl, and Li2O/Li2CO3 were weighed and mixed with a fixed molar ratio. The eutectic
mixture was first dried under vacuum for 48 h at 423 K to remove moisture completely. Ti2O3 was
prepared by the carbothermal reduction of TiO2 using activated carbon as the reducing agent at 1350 °C
[15–17]. Fig.1a indicates the XRD pattern of the sintered porous Ti2O3 electrode in which all peaks
visible correspond to Ti2O3. The result presents that there is no new phase yielded during the sintering
process. The SEM photograph of the sintered porous Ti2O3 electrode shown in Fig.1b. It suggests a
porous homogeneous structure. Porosity existed is favorable for the penetration of molten salts and the
diffusion of ionized O2− during electrolysis. The porosity of Ti2O3 electrode was estimated to be
60%~70% by the measurement of a mercury porosimetry (Auto Pore IV 9500, Micromeritics Instrument
Corp.). The porous product obtained from the carbothermal reduction was directly used as an electrode
in electrolysis experiments, the photo of that was illustrated as the inset of Fig.1b.
Int. J. Electrochem. Sci., Vol. 15, 2020
6111
Figure 1. XRD pattern (a) and SEM image (b) of porous Ti2O3 electrode used for electrolysis, the insets
in (b) is the photo of the porous Ti2O3 electrode.
2.2 Electrochemical methods
A programmable electrical resistance furnace equipped with a thermal controller was used for
heating purposes. Electrochemical experiments were performed using a three-electrode system. A 20-
mm-wide high-purity graphite sheet served as the counter electrode (CE). An electrode, composed of a
silver wire, 1 mm in diameter, dipped into a boron nitride tube containing a silver chloride solution with
a small hole giving access to the melt, was served as a Ag+/Ag reference electrode (RE). A tungsten
wire, 2 mm in diameter, was used as the working electrode (WE) for a blank experiment. Afterward, a
tungsten wire working electrode with a groove full of the Ti2O3 porous particles (W-Ti2O3 electrode)
was utilized to recognize the reduction mechanism. In addition, another two different types of working
electrode (WE) were used in the electrochemical procedure for studying the process of anode
polarization: one is a glassy carbon rod (2 mm in diameter), another is a Pt wire (1mm in diameter).
Cyclic voltammetry (CV) and other transient electrochemical techniques were performed using a
Princeton PARSTAT 2273 electrochemical workstation and a power amplifier (BOOSTER20A,
KEPCO).
The electrolysis process was carried out using a DC regulated power supply (IT6722A, ITECH)
and monitoring and acquisition of the electrical data were accomplished with the help of a data
acquisition system (Agilent 34401A) with a two-electrode cell. 464.5 g of LiCl-KCl-Li2O was
introduced in a high-pure graphite crucible and then placed in the sealed stainless steel chamber. A
porous Ti2O3 cathode connected with a tungsten wire was used as the cathode. Two kinds of electrolytic
processes with a graphite electrode (60 mm×80 mm) and a Pt electrode (40 mm×50 mm) as anode were
implemented separately under appropriate conditions. The mixture was kept under inert Ar flow at 5
L/min during its melting. After the temperature had been attained 800 °C, the porous Ti2O3 cathode was
slowly lowered into the electrolyte. The selected anode was then immersed in the molten salt. The power
supply was then connected to the anode and cathode connection leads. Continually-varied current
electrolysis on different values was performed with the corresponding voltage and electrode potential
measured. Electro-reduction was terminated and the electrodes were removed from the melt into the
upper part of the reactor. Then, the reactor turned cool. The salt adhering to the samples was removed
by water impregnation and the cleaned samples were dried under vacuum at room temperature.
(b)
(a)
Int. J. Electrochem. Sci., Vol. 15, 2020
6112
The cathode product was characterized by X-ray diffraction with a Cu-Kα characteristic ray
(XRD, X′ Pert Pro MPD, Philips) and scanning electron microscopy (SEM, S-4800, Hitachi). LECO
analyses (ON736 and SC832, LECO, Laboratory Equipment Corporation) were used for oxygen analysis
of the product. The lithium content in reduced samples was tested on Agilent ICP-MS 7900. The used
molten salt was dissolved in distilled water. After filtering the solution, excess BaCl2 solution (0.5
mol/L) was added to the filtered solution to estimate the CO32– content in the used electrolyte medium
by weighing the deposit.
3. RESULTS AND DISCUSSION
3.1 Thermodynamic considerations
Fig.2 shows the standard potentials vs. Li+/Li of possible cathode and anode reactions during the
electrochemical reduction of Ti2O3 in LiCl-KCl-Li2O melt at 800 °C [18]. For the anode process, reaction
7 (carbon consumption) is more favorable in thermodynamically when a graphite anode is employed.
Thus, the concentration of CO32– in molten salt would increase as the electrolysis proceeds. As a result,
the cathodic reaction graduates into the deposition of carbon (reaction 5) from the electrochemical
reduction of Ti2O3.
Figure 2. Standard potentials of the possible electrode reactions involved during electrolysis of Ti2O3 in
LiCl-KCl-Li2O melt at 800 °C using graphite and Pt electrode as the anode.
In theory, CO32– concentration in melt will achieve a dynamic equilibrium between carbon
deposition at cathode and carbon consumption at anode. However, as the cathode potential shifts
negatively and the anode potential shifts positively, the equilibrium could not be maintained for three
reasons: (i) more O2– brought from Ti2O3 at cathode would enter the molten salt; (ii) metallic Li would
Int. J. Electrochem. Sci., Vol. 15, 2020
6113
deposit at cathode; (iii) CO, CO2, and even Cl2 gas might be liberated at anode. Moreover, the complete
reduction of Ti2O3 to metallic Ti requires the cathodic potential approaching the deposition potential of
metallic Li from Fig.2. When a Pt anode is employed, it is found that O2 evolution is more favorable
thermodynamically. With the positive shift of potential, the dissolution of Pt metal occurs before Cl2
evolution. Concurrently, the cathodic reactions are mainly contained of stepwise reduction of Ti2O3 and
deposition of metallic Li with the negative shift of potential.
3.2 Electrochemical analysis
Figure 3. Cyclic voltammograms for (a) cathode and anode polarizations obtained on a different
electrode (cathode polarization-W working electrode, anode polarization-glassy carbon and Pt
working electrode) and (b) obtained on MCE of W-Ti2O3 in LiCl-KCl (46% LiCl, 54% KCl)
melt at 800 °C.
Fig.3a shows the typical CV curves obtained in molten LiCl-KCl at 800 °C. For investigating the
polarization process under the condition of stable electrode, cathodic and anodic polarization were
carried out on W, glassy carbon, and Pt electrode respectively. From the CV curve of cathodic
polarization obtained by using a W electrode, a sharp increase of cathodic current (wave c1) and the
corresponding anodic current (wave a1) were observed at around –1.9 V, which were attributed to the
deposition and dissolution (reaction 2 and 6, respectively) of metallic Li. Similarly, from the CV curve
of anodic polarization with Pt electrode, a sharp increase of the anodic current and the corresponding
cathodic current could been observed at about 1.1 V, which were brought from the dissolution and
deposition of Pt. When the glassy carbon was used as the working electrode, a current was noticed to
increase linearly with potential beyond 1.4 V, and it is a typical wave of the evolution of Cl2 gas (reaction
14).
As presented in Fig.3b, a series of normalized voltammograms was recorded in the LiCl-KCl
bath using the W-Ti2O3 electrode at 800 °C. The potential sweep ranging of the CV curves for studying
the reduction of Ti2O3 is from –0.2 to –2.3 V (vs. Ag+/Ag). Comparing with the CV curve of cathodic
polarization obtained by using a W electrode in Fig.3a, it indicates that wave c1 observed from Fig.3b
was corresponding to the deposition of Li. According to the calculated reduction potential of Ti2O3 and
TiO in Fig.2, waves c3 and c2 could represent the reduction of the Ti2O3 to TiO and TiO to metallic Ti,
(b)
(a)
Int. J. Electrochem. Sci., Vol. 15, 2020
6114
respectively. The electrochemical reduction of Ti2O3 can be summarized as Equation 15–18 by
considering LiTiO2 is an inevitable intermediate product during the electrochemical reduction of Ti2O3
in molten LiCl-KCl-Li2O. Accordingly, wave c3 was identified to correspond to reaction 15 and wave
c2 corresponded to reaction 16–18.
Ti2O3+Li++e→LiTiO2+TiO (15)
LiTiO2+e→TiO+Li++O2– (16)
LiTiO2+e→Ti+Li++O2– (17)
TiO+2e→Ti+O2– (18)
3.3 Electrolysis of Ti2O3 with a graphite electrode as anode
In order to inhibit the generation of Cl2 during the electrolysis process and elucidate the evolution
principle of C, O, and CO32–, the LiCl-KCl-Li2O and LiCl-KCl-Li2CO3 melts were examined by cyclic
voltammetry on a glassy carbon electrode at 800 °C. The result is shown in Fig.4. Seen from the CV
curve obtained in the LiCl-KCl-Li2O melt, two anodic waves (a2, a3) were observed. Referring to the
calculated discharge potentials of O2– in Fig.2, waves a2 and a3 should correspond to reactions 7 and 9.
In the case of anodic polarization in molten LiCl-KCl-Li2CO3, three anodic waves were observed.
Obviously, the wave a5 corresponds to the evolution of Cl2. The waves a1 and a4 were corresponding
to the reduction (reaction 5) and discharge (reaction 10 and 11) of CO32–, respectively [19]. In addition,
the discharge potential of CO32– is much more positive than that of O2– on glassy carbon electrode.
Figure 4. Cyclic voltammograms for anode polarization on a glassy carbon electrode in different melts
(LiCl: KCl: Li2O=45.6: 53.4: 1wt% and LiCl: KCl: Li2CO3=45.2: 52.8: 2 wt%) at 800 °C.
Continually-varied current electrolysis on different values (3 A, 4 A, 4.5 A and 5 A) with a plate
electrode of porous Ti2O3 using as the cathode was performed in molten LiCl-KCl-Li2O at 800 °C. The
relationships between time and voltage, cathodic and anodic potential are presented in Fig.5a.
Int. J. Electrochem. Sci., Vol. 15, 2020
6115
Figure 5. Electrolytic parameters versus time curves (a) for continually-varied current electrolysis on
different values using a graphite anode and XRD traces (b) of the reduced samples obtained after
electrolysis for different duration in LiCl-KCl-Li2O melt at 800 °C.
It observed that the voltage follows the current step. At the first stage, the cathodic potential
finally reached –1.07 V as the electrolysis continued on 3 A for 10 h, which is close to the deposition
potential of carbon (reaction 5). The starting anode potential was about 0.1 V, which approach to the
potential corresponding to the wave a2 in Fig.4. It shows that the starting reaction at anode could be
considered as the consumption of carbon anode (reaction 7). The positive shift of both anodic and
cathodic potentials at the first stage indicated that the content of active ions (O2– and CO32–) in the molten
salt might have changed. It can be known from previous electrochemical analysis, the discharge potential
of O2– ions (reaction 7) is much more positive than that of CO32– ions (reaction 10 and 11) on a glassy
carbon electrode. Furthermore, the discharge potential of CO32– ions (reaction 5) at the cathode is much
more positive than the deposition potential of Li+ and the electrochemical reduction potential of TiO and
LiTiO2. As a result, both the anodic and cathodic potentials shifted positively with O2– ions decreased
and CO32– ions increased in melts at the initial stage of electrolysis. From the electrolytic current
increases to 4 A, the cathodic potential became more and more negative and the anodic potential became
more and more positive with the increasing of electrolytic current. When raising the electrolytic current
to 5 A, the cathodic potential could achieve on –1.95 V. It proved that the deposition potential of metallic
Li had been reached at 800 °C.
Fig.5b shows the XRD traces of the product obtained after continually-varied current electrolysis
for different duration. It indicates that TiO and LiTiO2 as intermediate products could be obtained during
the continually-varied current electrolysis progress. Further increase the electrolytic current and extend
the time, T3O and TiC could be identified. The formation of Ti3O and TiC could be attributed to the
following reactions:
TiO+CO32–+6e→TiC+4O2– (19)
LiTiO2+CO32–+8e→TiC+Li++5O2– (20)
3TiO+4e→Ti3O+2O2– (21)
3LiTiO2+10e→Ti3O+3Li++5O2– (22)
(b)
(a)
Int. J. Electrochem. Sci., Vol. 15, 2020
6116
Finally, the product of Ti with a small amount of TiC doped was obtained after continually-
varied current electrolysis for 47 h. It can be inferred that a little TiC formed on the surface of samples
due to CO32– ions discharged on the surface of the cathode.
Figure 6. SEM images and EDS analysis results of the reduced samples obtained after continually-varied
current electrolysis for 10 h (a), 19 h (b), 29 h (c) and 47 h (d) in LiCl-KCl-Li2O melt at 800 °C.
The SEM images and EDS analysis results of cathodic products obtained after continually-varied
current electrolysis for different duration are shown in Fig.6. The LiTiO2 and TiO particles can be
distinguished clearly seen from Fig.6a. After continually-varied current electrolysis on 4 A for 9 h,
product morphology had no changed significantly seen from Fig.6b. A porous structure was observed
from Fig.6c. It was identified as the typical structure of sintered Ti3O according to the XRD pattern in
Fig.5b. Fig.6d shows the SEM image of the reduced sample after 47 h of electrolysis. Ti particles
presented scalariform.
(b)
(a)
(d)
(c)
Int. J. Electrochem. Sci., Vol. 15, 2020
6117
Figure 7. The Li/CO32– contents in the cathodic product and melt concerning the different electrolysis
times during continually-varied current electrolysis using a graphite electrode in LiCl-KCl-Li2O
melt at 800 °C, respectively.
From thermodynamic analysis and electrochemical tests, the electrolysis system contains a
balance of carbon and oxygen. The carbon anode was oxidized and combined with O2– ions to form
CO32–. When CO3
2– accumulates enough, it will discharge at cathode to form C and O2– ions. Obviously,
the amount of the CO32– depends on that of O2– ions in molten salt. Because testing the concentration of
O2– ions was difficult, we mainly tested the CO32– ion in molten salt. In order to understand the
electrochemical reduction process, the Li content in cathode at different stages was also detected. The
results are plotted in Fig.7. It can be seen that the Li content in cathode and the concentration of CO32–
in melt increased greatly in the first stage of constant current electrolysis with 3 A for 10 h. Therefore,
the main cathodic reaction in the initial stage was the intercalation of Li+ ions into the porous cathode,
and the main anodic reaction was the consumption of carbon anode. After raising the electrolytic current,
the Li content in the reduction product decreased slightly, which indicated LiTiO2 was reduced as the
electrolysis proceeds. LiTiO2 had decomposed completely after electrolysis of 47 h. At this time, there
is no Li in the product. Comparing with Li content in the porous cathode, there was no significant change
in the content of CO32– ions in melt after increasing the electrolytic current during the electrolysis. It can
be deduced that CO32– ions always were in dynamic balance in the molten salt during the subsequent
electrolysis. Further, the small change in the concentration of CO32– can also come from the addition of
Li2O in the molten salt. As a result of giving rise to CO32–, it in turn could adversely affect the Li2O
recycling necessary for reduction of the oxide. Many studies suggest that graphite anode is also relatively
less stable in low-melting LiCl-base melt than in CaCl2 melt [20]. For these reasons, the graphite anode
was replaced with a Pt electrode for the electrolysis.
3.4 Electrolysis with a Pt electrode
Carbon consumption and participation in electrode reactions lead to an extremely inefficient
electrolysis process using a graphite electrode. Therefore, a more stable Pt anode was chosen for studying
the continually-varied current electrolysis process of the porous Ti2O3 cathode.
Int. J. Electrochem. Sci., Vol. 15, 2020
6118
Fig.8 shows the cyclic voltammograms obtained during anodic polarization of a Pt wire in LiCl-
KCl-Li2O molten salt at 800 °C.
Figure 8. Cyclic voltammograms for anode polarization during different sweep range on Pt working
electrode in LiCl-KCl-Li2O melt at 800 °C.
Figure 9. Electrode potentials versus time curves (a) for electrochemical reduction experiments using a
Pt anode in LiCl-KCl-Li2O melt at 800 °C; the surface appearance photos (b) of Pt anode before
and after the electrolysis; the XRD spectrum (c) and SEM image (d) of the reduction product
obtained after continually-varied current electrolysis for 26 h.
(b)
(a)
(d)
(c)
Int. J. Electrochem. Sci., Vol. 15, 2020
6119
Three anodic wave A1, A2, and A3 and one cathodic wave C1 could be observed in Fig.8. It has
been confirmed in previous studies that anodic wave A1, A2 and A3 correspond to the formation of
Li2PtO3 coating (2Li+ + Pt + 3O2−→ Li2PtO3 + 4e), O2 gas evolution (O2− → 1/2O2 + 2e) and the
dissolution of the Pt wire (Pt → Pt2+ + 2e) respectively [21, 22]. The cathodic wave C1 corresponds to
the decomposition of the Li2PtO3 coating (Li2PtO3 + 4e→2Li+ + Pt + 3O2−). Compared with the
dissolution potential (1.1 V) of the Pt electrode in LiCl-KCl molten salt, that in LiCl-KCl-Li2O molten
salt shifted positively to around 1.3 V because a Li2PtO3 inoxidizing coating formed. However, wave
C1 had weakened when the anode limit is more positive and reached 1.5 V. It indicated that the Li2PtO3
coating had been dissolved or exfoliated from the surface of the Pt electrode. As a result, the anode
potential should be less positive than 1.3 V when a Pt sheet was used as the anode for electrolysis.
The relationships between time and electrode potentials reflect to the continually-varied current
electrolysis carried out with a porous Ti2O3 cathode and a Pt anode in molten LiCl-KCl-Li2O at 800 °C
are presented in Fig.9a. In order to avoid the deposition of an amount of metallic Li at cathode, the
cathode potential was monitored by turning down the electrolytic current gradually. When the
electrolysis with 1 A for 4 h, the cathodic potential was around –1.3 V and began to have a negative
shift. After electrolysis for 7 h, the cathodic potential (–2.0 V) had exceeded the deposition potential of
metallic Li accompanying the electrolytic current down to 0.5 A. Accordingly, the cathodic potential
shifted positively to around –1.8 V. As the electrolysis continued, the cathodic potential shifted
negatively to 2.0 V after electrolysis for about 15 h again. The current was further turned down to 0.25
A.
Fig.9b shows the photos of Pt anode before and after the electrolysis. It is obviously observed
that a yellow coating was formed on the surface of the Pt electrode after electrolysis, which had been
identified to be Li2PtO3. Fig.9c shows the XRD pattern of the cathodic reduction product obtained after
continually-varied current electrolysis for 26 h. The crystal structure of pure titanium was detected. The
morphology of the cathodic reduction product was analyzed and shown in Fig.9d. A porous structure
was found. The oxygen content of the cathode product was measured and the result showed that it was
approximately 4.6%.
3.5. Discussion
Contrast with CaCl2 electrolyte system, using LiCl electrolyte system has three advantages [23]:
(i) The current efficiency is much better because the LiCl melt could easier permeate into the sample.
Especially the porous Ti2O3 electrode was used in this work. (ii) Lower melting point, lower energy
consumption. (iii) Avoiding the formation of refractory Ca2TiO3. However, only Ti-O solid solution as
electrochemical reduction product can be obtained in LiCl-based molten salt. While in CaCl2-based
molten salt, titanium metal containing as low as 0.3% oxygen can be prepared [4, 24]. Fig.10a and b
show the E-pO2–diagrams of Ti-O-Cl-Li and Ti-O-Cl-Ca systems respectively [8, 25, 26]. From the
diagrams, to obtain a Ti-O solid solution containing 1% oxygen as an example, it could be achieved in
molten CaCl2 saturated with CaO effortlessly. Whereas, a product of the same quality could be obtained
with the activity of O2– lower than 1.8×10–2 in molten LiCl requires. Thus, low-oxygen titanium was not
Int. J. Electrochem. Sci., Vol. 15, 2020
6120
prepared in LiCl-based molten salt could be due to the relatively higher O2– ion activity in the molten.
The weaker reducibility of lithium than calcium may also be responsible for the high oxygen content of
titanium products. It suggests that such an electrolytic process could be attractive for the production of
Ti crude product with slightly high oxygen content [27].
Figure 10. Diagrams of the equilibrium potential E (relative to the standard chlorine electrode) against
the activity of oxide ion (expressed as its logarithm, pO2–) in the systems of Ti-Cl-O-M.
5. CONCLUSION
In this work, titanium metal could be prepared via continually-varied current electrolysis of
porous Ti2O3 electrode in molten LiCl-KCl-Li2O at 800 °C with a graphite and a Pt anode, respectively.
During the reduction process from Ti2O3 to Ti by electrolysis using a graphite anode, the Ti2O3 cathode
undergoes several intermediate phases, such as LiTiO2, TiO, Ti3O, and finally metal Ti doped with TiC
could be obtained. Carbonless titanium could be obtained via continually-varied current electrolysis for
26 h using a Pt anode. During electrolysis with a Pt electrode, a Li2PtO3 coating layer could be formed
on the surface of Pt electrode at 0.5 V, which makes the dissolution potential of Pt move forward and
play a protective role. In addition, the current efficiency increases greatly by the application of an inert
Pt anode instead of graphite anodes. Therefore, developing a low-cost inert anode material is essential
for the production of titanium by electrolysis in future. Comparing to prepare titanium in CaCl2-based
molten salt, the titanium product obtained in LiCl-based melt has a high oxygen content.
ACKNOWLEDGEMENTS
This research work was supported by Science and Technology Program of Chongqing, China (grant
number cstc2018jcyjAX0833), and Fuling District Science and Technology Plan Project of Chongqing,
China (grant number FLKJ2018BBA3075).
References
1. G.Z. Chen and D.J. Fray, Extractive Metallurgy of Titanium, (2020) 227.
2. C. Wu, M. Tan, G. Ye, D.J. Fray and X. Jin, ACS Sustainable Chem. Eng., 7 (2019) 8340.
(b)
(a)
Int. J. Electrochem. Sci., Vol. 15, 2020
6121
3. M.R. Earlam, Extractive Metallurgy of Titanium, (2020) 97.
4. G.Z. Chen, D.J. Fray and T.W. Farthing, Nature, 407 (2000) 361.
5. C. Schwandt and D.J. Fray, Electrochim. Acta, 51 (2005) 66.
6. K. Jiang, X. Hu, H.J. Sun, D. Wang, X. Jin, Y. Ren and G.Z. Chen, Chem. Mater., 16 (2004) 4324.
7. H.S. Shin, J.M. Hur, S.M. Jeong and K.Y. Jung, J. Ind. Eng. Chem., 18 (2012) 438.
8. K. Dring, R. Dashwood and D. Inman, J. Electrochem. Soc., 152 (2005) E104.
9. D.T.L. Alexander, C. Schwandt and D.J. Fray, Acta Mater., 54 (2006) 2933.
10. K. Jiang, X. Hu, M. Ma, D. Wang, G. Qiu, X. Jin and G.Z. Chen, Angew. Chem.Int. Ed., 45 (2006)
428.
11. C. Schwandt, G.R. Doughty and D.J. Fray, Key Engineering Materials, 436 (2010) 13.
12. D.S. Maha Vishnu, N. Sanil, L. Shakila, R. Sudha, K.S. Mohandas and K. Nagarajan, Electrochim.
Acta, 159 (2015) 124.
13. R. Koc, J. Mater. Sci., 33 (1998) 1049.
14. A.A. Valeeva, S.Z. Nazarova and A.A. Rempel, J. Alloys Compd., 817 (2020) 153215.
15. Y.C. Woo, H.J. Kang and D.J. Kim, J. Eur. Ceram. Soc., 27 (2007) 719.
16. W. Sen, H.Y. Sun, B. Yang, B.Q. Xu, W.H. Ma, D.C. Liu and Y.N. Dai, Int. J. Refract. Met. Hard
Mater., 28 (2010) 628.
17. W. Sen, B.Q. Xu, B. Yang, H.Y. Sun, J.X. Song, H.L. Wan and Y.N. Dai, Trans. Nonferrous. Met.
Soc. China, 21(2011)185.
18. J.M. Hur, J.S. Cha and E.Y. Choi, ECS Electrochem. Lett., 3 (2014) E5.
19. J. Ge, S. Wang, L. Hu, J. Zhu and S. Jiao, Carbon, 98 (2016) 649.
20. K.S. Mohandas, Trans. Inst. Min. Metall., Sect. C, 122 (2013) 195.
21. S.M. Jeong, H.S. Shin, S.H. Cho, J.M. Hur and H.S. Lee, Electrochim. Acta, 54 (2009) 6335.
22. T.B. Joseph, N. Sanil, L. Shakila, K.S. Mohandas and K. Nagarajan, Electrochim. Acta, 139 (2014)
394.
23. Y. Sakamura, M. Kurata and T. Inoue, J. Electrochem. Soc., 153 (2006) D31.
24. I. Park, T. Abiko and T.H. Okabe, J. Phys. Chem. Solids, 66 (2005) 410.
25. R. Littlewood, J. Electrochem. Soc., 109 (1962) 525.
26. T.H. Okabe, R.O. Suzuki, T. Oishi and K. Ono, Mater. Trans., JIM, 32 (1991) 485.
27. K. Dring, Key Engineering Materials, 436 (2010) 27.
© 2020 The Authors. Published by ESG (www.electrochemsci.org). This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).